Another Look at Forecast Accuracy Metrics for Intermittent Demand

TLDR

Traditional forecast accuracy metrics can yield infinite or undefined values for intermittent demand data. Rob Hyndman reviews existing forecast accuracy metrics and discusses their shortcomings. He introduces the mean absolute scaled error (MASE) as a more suitable metric for intermittent demand. He advocates adopting MASE as the standard metric for comparing forecast accuracy across multiple time series. © 2006 International Institute of Forecasters.

Abstract

Some traditional measurements of forecast accuracy are unsuitable for intermittent demand data because they can give infinite or undefined values. Rob Hyndman summarizes these forecast accuracy metrics and explains their potential failings. He also introduces a new metric-the mean absolute scaled error (MASE)-which is more appropriate for intermittent-demand data. More generally, he believes that the MASE should become the standard metric for comparing forecast accuracy across multiple time series. Copyright International Institute of Forecasters, 2006